On Sunday, the internet stopped working for millions of users worldwide.
But while many probably tried to restart their routers and unplug the power plug and plug it back in, it turned out that the internet problems started much further away.
The bug started in Canada
The error can be traced all over the Atlantic, more specifically a data center owned by CenturyLink in Mississauga, a million-strong city outside Toronto, Canada, writes the website ZDNet.
CenturyLink is a giant internet provider in North America, which had problems on Sunday afternoon Norwegian time. The problems were related to their firewall and so-called BGP routing. BGP routing is described by ZDNet as «the glue that keeps the Internet up», because it is messages that ISPs send to each other, telling what kind of IP addresses are available on their network.
An error in a so-called Flowspec at CenturyLink is said to have caused them to send out incorrect BGP routes to other major internet providers. This created a domino effect that spread all over the world.
Many pages were down
After seven hours, the error at CenturyLink had been resolved. Among other things, it must have involved asking all of its so-called «Tier 1» partners to ignore incoming traffic from their servers, and to restart all equipment, so that they could start again with a «clean» BGP list.
In addition to creating problems for many Europeans, it also put many American websites out of play. Servers at Amazon, Twitter, Microsoft (Xbox Live), EA, Blizzard, Steam, Discord, Reddit, Hulu, OpenDNS and many others had issues on Sunday as a result of the CenturyLink loss.